We found an increased risk of neuropsychiatric disorders in children born with FIRS compared to those without FIRS. More specifically, there is increased incidence of ASD, ADHD, conduct disorder, PTSD, and any psychiatric disorder in these children, which was significant even after adjusting for maternal history of psychiatric disorders, maternal substance use during pregnancy, and maternal lifetime history of suicide attempts. While maternal alcohol, amphetamine, cannabis use disorder, and neonatal substance use disorder diagnoses were significantly increased in children born with FIRS compared to those without, children born with FIRS continued to show increased risk of psychiatric disorders even after adjusting for these potential confounders. These results suggest that inflammation of the placenta has a critical role in the pathogenesis of neuropsychiatric disorders during childhood, independent of many potential confounders.

We discuss these results in the context of its strengths and limitations. This study is the first to examine the impact of FIRS on risk for neuropsychiatric disorders using an EHR cohort. We included children born up to 2018 and had at least 5 years of follow-up and focused on childhood onset diagnoses and as such our results are not generalizable to adolescent onset diagnoses. It is also possible that some patients may be receiving care outside of the UPMC system and as such, we were not able to capture their offspring’s medical records. However, the region of this patient population is dominated by one child and adolescent psychiatric care system, reducing the possibility of patients being lost to follow up entirely, or follow up at other institutions. While the standard of practice at the women’s hospital used in the study is to only examine placentas which meet certain clinical criteria (e.g., low birth weight, intrauterine growth restriction, complications at delivery, poor newborn health), it is possible that abnormal placentas existed in the control group by virtue of not being examined microscopically. A recent study demonstrates that while many placentas have some level of inflammatory changes, they are almost always minor and unrelated to clinical outcomes [37]. In one study, only 6.7% of placentas assumed to be normal met criteria for FIRS upon closer microscopic examination [33]. Thus, the misclassification of FIRS placentas in the control group is likely to be small. We attempted to control for anti-inflammatory medications that may have inadvertently treated inflammation caused by FIRS. Additionally, some medications have both anti-infection and anti-inflammatory properties (e.g., rifampin), which could have treated FIRS (if caused by infectious etiology) and/or subsequent inflammation. We only considered maternal history of psychiatric disorders and were not able to account for paternal history.

Considering limitations related to EHR psychiatric outcomes, the DSM was revised from IV-TR to V during the study period. Liberalization and refining of some criteria, such as ASD, could have resulted in changes in the prevalence of diagnoses across the study period. However, it is likely that this refinement resulted in more accurate diagnosis, reflecting a possible underestimate of the true prevalence of these disorders [38]. Cultural changes occurred undoubtably and could have effects on parenting, social programs, and the societal milieu of children that were not captured in this study; however, these affected the groups of children with and without FIRS. We also had no information about patient’s history of trauma, given its lack of reliable ICD coding.

Considering the pathologic limitations to this study, the gold standard for FIRS diagnosis is cord blood (plasma Il-6 in fetal cord blood must be greater than 11 ng/ml); however, cord blood was not available for analysis in this study. Histologically, FIRS can be diagnosed by the presence of funisitis or chorionic vasculitis. While the criteria have been consistent throughout the study period (and since its inception as a diagnostic entity), this study did not account for the severity of fetal placental inflammation. Despite subspecialized gynecologic pathologists reviewing these records, there are variations among diagnoses that are rendered or emphasized for placental pathology reports, often with emphasis on clinically actionable data, such as chorioamnionitis. There is also the possibility that some diagnoses were placed in pathology report comment, and thus not captured using diagnostic codes. Funisitis and chorionic vasculitis are not completely specific for fetal inflammation, as both can be seen in other conditions such as chorioamnionitis.

We had no information about maternal infection during pregnancy and maternal stress in utero, which could result in increased maternal inflammation and immune responses. Direct and indirect evidence suggest that maternal immune activation is associated with increased risk of neuropsychiatric disorders. A Swedish registry study showed that fetal exposure to maternal infection increased the risk for autism and depression, but not bipolar or psychotic disorders [39]. Maternal stress in utero has known associations with psychiatric disease in offspring [40]. Maternal infections have also been directly associated with increased risk for ASD and schizophrenia in the offspring [41,42,43]. Maternal diseases that are associated with increased immune responses (e.g., autoimmune disorders, asthma) and maternal exposure to environmental and psychological stressors were previously found to be associated with increased risk for ASD and schizophrenia [40]. Maternal immune activation has been shown to affect such risk in the offspring through alterations in the prenatal environment, specifically through dysregulations in inflammatory pathways [44]. Animal models showed that IL-6 can directly cross the placenta and target the fetal brain and that a single injection of IL-6 in pregnant mice resulted in reduced social interaction and other behavioral abnormalities characteristic of ASD [45, 46]. Additionally, systemic maternal inflammation during pregnancy, especially elevated Il-6 levels, have been associated with cognitive impairment and development of psychiatric disease [47,48,49,50]. IL-1β was found to directly affect brain development through an alteration of the proliferation and differentiation of neural progenitor cells (NPCs) towards increasing gliogenesis and reducing neurogenesis [50]. Similarly, it has been shown that elevated maternal cytokines were associated with dendritic abnormalities characteristic of schizophrenia [9]. Indeed, increased IL-6, TNF-α, and other pro-inflammatory cytokines have been consistently implicated in psychiatric disorders, specifically, ASD, schizophrenia, and bipolar disorders [51,52,53,54]. Therapeutically, ongoing trials are using IL-6 inhibitors in pregnant women to treat FIRS in utero. Screening for FIRS and other risk factors during prenatal care is important to reduce the risk of childhood psychiatric disease.

Additional biological and molecular mechanisms are theorized to include the downstream effects of inflammation at a critical junction in neurodevelopment. As FIRS occurs later than 20 weeks of gestation, it occurs during a time of heightened vulnerability of prenatal brain development to environmental factors [20,21,22,23]. Alterations in epigenetic gene expression [55,56,57] could be an underlying mechanism by which FIRS (and its subsequent inflammation), alters the risk profile of a developing child. However, genetic factors could be also implicated in FIRS. Indeed, GWAS studies of ASD and ADHD show a shared genetic etiology along with schizophrenia and major depressive disorder [58] and the implicated genetic loci across these childhood- and adult-onset disorders were enriched in neuronal and immune pathways [59].

To our knowledge, this is the first study to examine fetal inflammatory syndrome as determined by pathological reports and its relationship with risk and onset of neuropsychiatric disorders. Children born with placenta meeting criteria for FIRS are at higher risk for neuropsychiatric disorders compared to children without FIRS. This study has potential implications for clinical care and prevention approaches. Children born with placenta meeting criteria for FIRS should be monitored closely for early identification and treatment. Future studies are needed to identify the biological mechanisms through which inflammation in general and FIRS in specific increases risk for neuropsychiatric disorders.